@tduncan / Nvidia-Optimization-Stack-VLLM
Code Commits Issues Pulls Cloud Agent CI/CD
main
main
yaml 10 lines 250 Bytes
Raw
1 model_name: "RedHatAI/Qwen3.6-35B-A3B-NVFP4"
2 accuracy_threshold: 0.91
3 num_questions: 1319
4 num_fewshot: 5
5 gen_prefix: " <think>\n\n</think>\n"
6 server_args: >-
7 --enforce-eager
8 --max-model-len 8192
9 --moe-backend humming
10 --linear-backend humming